Added generic roles

sd-nlp-non-tokenized.py

@@ -70,14 +70,15 @@ class SourceDataNLP(datasets.GeneratorBasedBuilder):
     VERSION = datasets.Version("1.0.0")
 
     _URLS = {
-        "NER": f"{_BASE_URL}sd_panels_filtered.zip",
-        "PANELIZATION": f"{_BASE_URL}sd_panelization_filtered.zip",
+        "NER": f"{_BASE_URL}sd_panels_filtered_generic_roles.zip",
+        "PANELIZATION": f"{_BASE_URL}sd_panelization_filtered_generic_roles.zip",
     }
     BUILDER_CONFIGS = [
         datasets.BuilderConfig(name="NER", version=VERSION, description="Dataset for entity recognition"),
         datasets.BuilderConfig(name="GENEPROD_ROLES", version=VERSION, description="Dataset for semantic roles."),
         datasets.BuilderConfig(name="SMALL_MOL_ROLES", version=VERSION, description="Dataset for semantic roles."),
         datasets.BuilderConfig(name="BORING", version=VERSION, description="Dataset for semantic roles."),
+        datasets.BuilderConfig(name="ROLES", version=VERSION, description="Dataset to train roles. GENEPROD and SMALL_MOL at once."),
         datasets.BuilderConfig(
             name="PANELIZATION",
             version=VERSION,
@@ -147,6 +148,16 @@ class SourceDataNLP(datasets.GeneratorBasedBuilder):
                     "tag_mask": datasets.Sequence(feature=datasets.Value("int8")),
                 }
             )
+        elif self.config.name == "ROLES":
+            features = datasets.Features(
+                {
+                    "words": datasets.Sequence(feature=datasets.Value("string")),
+                    "labels": datasets.Sequence(
+                        feature=datasets.ClassLabel(num_classes=len(self._PANEL_START_NAMES),
+                                                    names=self._PANEL_START_NAMES)
+                    ),
+                }
+            )
 
         return datasets.DatasetInfo(
             description=self._DESCRIPTION,
@@ -161,7 +172,7 @@ class SourceDataNLP(datasets.GeneratorBasedBuilder):
         """Returns SplitGenerators.
         Uses local files if a data_dir is specified. Otherwise downloads the files from their official url."""
 
-        if self.config.name in ["NER", "GENEPROD_ROLES", "SMALL_MOL_ROLES", "BORING"]:
+        if self.config.name in ["NER", "GENEPROD_ROLES", "SMALL_MOL_ROLES", "BORING", "ROLES"]:
             url = self._URLS["NER"]
             data_dir = dl_manager.download_and_extract(url)
             data_dir += "/"
@@ -230,6 +241,9 @@ class SourceDataNLP(datasets.GeneratorBasedBuilder):
                 elif self.config.name == "BORING":
                     yield id_, {"words": data["words"],
                                 "labels": data["label_ids"]["boring"]}
+                elif self.config.name == "ROLES":
+                    yield id_, {"words": data["text_generic_roles"],
+                                "labels": data["labels_generic_roles"]}
                 elif self.config.name == "PANELIZATION":
                     labels = data["label_ids"]["panel_start"]
                     tag_mask = [1 if t == "B-PANEL_START" else 0 for t in labels]